Abstract
Thromboembolic disease is a frequent complication in cancer. Tissue factor (TF), shown to be involved in tumor growth and metastasis, is also considered to play a central role in the pathogenesis of cancer-associated thrombosis. Circulating TF-bearing microparticles (TF+ MPs) have been found in the plasma of patients with different malignancies and are thought to contribute to their hypercoagulable state. Although numerous studies have focused on TF+ MPs derived from blood cells, there is no information available on the pathological relevance of MPs originating from tumor cells. We conducted a study to detect, enumerate and characterize the procoagulant activity (PCA) of MPs released specifically from tumor cells. MPs from high (MDA-231) and low (MCF-7) TF-expressing human breast carcinoma cells were generated ex vivo in whole blood or in buffer under stirring conditions for 45 minutes. The numbers (MPs/ml) of total and TF-expressing tumor-derived particles (TMPs) in cell-free plasmas were measured by flow cytometry using FITC-labeled annexin V and a PE-labeled monoclonal anti-TF antibody respectively. The PCA of TMPs was measured by a one stage clotting assay and a highly sensitive fluorogenic thrombin generation assay. In order to evaluate the PCA of circulating TMPs, we injected 2x106 TF+ MPs derived from MDA-231 cells into mice via the tail vein. Human TF antigen and activity were measured in cell-free mouse plasmas at various intervals (5–420 min) after injections by ELISA and clotting assay, respectively. MPs less than 1μm in diameter were released from tumor cells in both buffer and whole blood by stirring. TMPs positive for TF consisted of approximately 40% of the annexin V+ MPs, and such particles derived from as low as 1x105 MDA-231 cells could be enumerated reliably (2.5x104 MPs/105 cells). By ultracentrifugation of cell-free plasmas, we confirmed that TF antigen was associated entirely with the MP fraction. TMPs derived from as few as 450 MDA-231 cells shortened plasma recalcification times from 525 ± 114 to 265 ± 148 (P<0.01), and significantly accelerated thrombin generation as evidenced by a 3 fold shortening in lag time, and a 2 fold increase in the rate of thrombin formation and thrombin concentration. No PCA was detected with MCF-7-derived TMPs. The PCA of TMPs was inhibited completely by a blocking anti-TF monoclonal antibody, but not by saturating concentrations of annexin V (an inhibitor of phospholipid PCA) or corn trypsin inhibitor (an inhibitor of the intrinsic pathway). Five minutes following the injection of TMPs into mice, appreciable levels of human TF antigen and activity were detected in cell-free plasmas. Both TF activity and antigen decreased over time and were detectable no longer than 30 minutes after injection, indicating a rapid clearance of circulating TMPs in vivo. In contrast, when TMPs were incubated in human whole blood ex vivo for various intervals, TF activity remained unchanged in cell-free plasmas for at least 5 hrs and TF antigen was not detected by flow cytometry on any blood corpuscles, including platelets, at all intervals. However, when whole blood containing TMPs was clotted by recalcification, no TF activity could be detected in the serum, indicating the incorporation of TMPs in formed clots. In summary, MPs bearing active TF are released readily from tumor cells and can significantly activate coagulation even at very low concentrations. These results provide new insights towards the potential contribution of TMPs to the pathogenesis of cancer-associated thrombosis.
Disclosure: No relevant conflicts of interest to declare.
Author notes
Corresponding author